How significant is the Pistons’ 3-0 start?

Mar 18, 2015; Philadelphia, PA, USA; Detroit Pistons head coach Stan Van Gundy in a game against the Philadelphia 76ers at Wells Fargo Center. The 76ers won 94-83. Mandatory Credit: Bill Streicher-USA TODAY Sports
Mar 18, 2015; Philadelphia, PA, USA; Detroit Pistons head coach Stan Van Gundy in a game against the Philadelphia 76ers at Wells Fargo Center. The 76ers won 94-83. Mandatory Credit: Bill Streicher-USA TODAY Sports /
facebooktwitterreddit

If you’re on this site reading this article, you probably know that the Detroit Pistons are 3-0. And that such a hot start would not have been predicted by almost any objective observers (nor by me, whether I qualify as objective or not). Being one of the NBA’s only six undefeated teams is exciting, but is it enough to revise expectations for the team?

To help answer this question, I’m looking at how teams that start 3-0 tend to fare. If the NBA were entirely random and all these outcomes were complete flukes, we’d expect 12.5% of all teams to start 3-0, for half of them to lose their fourth game, half the remainder to lose their fifth game, and so on. They would go .500 the rest of the way and, on average, end their seasons 42.5-40.5.

If it is a symptom of real superior performance, we would expect to see more teams start out with such a record, fewer than half of them lose each ensuing game, and the average such team to win more than 42.5 games. If the opposite happened, that would be weird and would suggest that somehow starting hot jinxes a team. If instead, the scenario described in the previous paragraph were to unfold, we’d have to conclude those hot starts were complete flukes and had no predictive bearing whatsoever.

Here is a complete list of every team to be unbeaten through three games for the last 15 seasons:

SeasonTeamInitial Win StreakFinal RecordProrated Total Wins
2000-01Cleveland Cavaliers430-5230
2000-01Philadelphia 76ers1056-2656
2000-01San Antonio Spurs358-2458
2000-01Utah Jazz553-2953
2001-02Los Angeles Lakers758-2458
2001-02Milwaukee Bucks441-4141
2001-02Minnesota Timberwolves650-3250
2001-02Sacramento Kings461-2161
2002-03Dallas Mavericks1460-2260
2002-03New Jersey Nets449-3349
2002-03Sacramento Kings359-2359
2002-03Seattle SuperSonics440-4240
2003-04Los Angeles Lakers556-2656
2003-04New Orleans Hornets341-4141
2003-04Seattle SuperSonics337-4537
2004-05Dallas Mavericks458-2458
2004-05Indiana Pacers444-3844
2004-05Miami Heat459-2359
2004-05Phoenix Suns462-2062
2004-05Toronto Raptors333-4933
2004-05Utah Jazz426-5626
2005-06Detroit Pistons864-1864
2005-06Los Angeles Clippers347-3547
2005-06Milwaukee Bucks340-4240
2005-06Washington Wizards342-4042
2006-07Los Angeles Lakers342-4042
2006-07NO/OKC Hornets439-4339
2006-07Philadelphia 76ers335-4735
2006-07Utah Jazz451-3151
2007-08Boston Celtics866-1666
2007-08Detroit Pistons359-2359
2007-08Houston Rockets355-2755
2007-08Indiana Pacers336-4636
2007-08Los Angeles Clippers423-5923
2007-08New Orleans Hornets456-2656
2007-08San Antonio Spurs356-2656
2008-09Atlanta Hawks647-3547
2008-09Detroit Pistons439-4339
2008-09Houston Rockets353-2953
2008-09Los Angeles Lakers765-1765
2008-09New Orleans Hornets349-3349
2008-09Toronto Raptors333-4933
2009-10Boston Celtics650-3250
2009-10Denver Nuggets553-2953
2009-10Miami Heat347-3547
2009-10Orlando Magic359-2359
2009-10Phoenix Suns454-2854
2009-10Utah Jazz553-2953
2010-11Atlanta Hawks644-3844
2010-11Los Angeles Lakers857-2557
2010-11New Orleans Hornets846-3646
2010-11Portland Trailblazers348-3448
2011-12Atlanta Hawks340-2649.7
2011-12Indiana Pacers342-2452.2
2011-12Miami Heat546-2057.2
2011-12Oklahoma City Thunder547-1958.4
2011-12Portland Trailblazers328-3834.8
2012-13New York Knicks654-2854
2012-13San Antonio Spurs458-2458
2013-14Houston Rockets354-2854
2013-14Indiana Pacers956-2656
2013-14Minnesota Timberwolves340-4240
2013-14Philadelphia 76ers319-6319
2014-15Golden State Warriors567-1567
2014-15Houston Rockets656-2656
2014-15Memphis Grizzlies655-2755
2014-15Miami Heat337-4537

There have been 446 teams over those seasons (the NBA expanded to 30 in 2004-05). The first thing to note is how many such teams there have been. If all teams were equal and there was nothing else to go on, we’d expect 56 3-0 teams. 67 is more than that, but not all that much. That’s about the number you’d expect if the range of the majority of winning percentages from the best to worst teams were about 42%-58%. In other words, if 90% of teams ended up with 34-48 wins. So that’s odd. Chalk one up for it meaning something to have a hot start, but not as much as might be expected.

On the other hand, 59.7% of these teams went on to win a fourth straight game. 34.3% won a fifth, 23.9% won a sixth, and 13.4% won a seventh. After that, the number of teams in this sample size is small enough to make it pretty useless for gathering conclusions. Teams went on to win almost exactly 49 games on average (prorating the lockout shortened 2011-12 season). All of those numbers are nearly perfectly consistent with .600 teams being the typical variety to start 3-0. That’s encouraging.

3-0teams
3-0teams /

It shouldn’t be shocking if the Pistons end up with a win total in the 30s. 15% of 3-0 teams do (with another 5% even lower). A team with underwhelming expectations out the gate that won relatively close games would be the variety you’d expect to achieve that outcome. On the other hand, the Pistons haven’t beat up on what shapes up (again, based on minimal information) to be the dregs of the league. Their three victims (Atlanta, Utah, and Chicago) are a combined 8-0 otherwise.

If I had to revise my 37-45 prediction today, it would probably be to about 41-41. That maybe overreacting, but hopefully it turns out to be way too conservative. We’ll have to see what happens in the upcoming matches with Indiana and Phoenix. Teams that started 4-0 won 52 games on average and teams that started 5-0 bumped that all the way to 56 wins. After that, any remaining returns for lengthening the opening win streak drop to nearly zero, at least based on historical precedent.